Page 521 - Advanced engineering mathematics
P. 521

14.7 DFT Approximation of the Fourier Transform  501


                                        Using these numbers, we can compute
                                                           127
                                                                 πik/2
                                                              V k e
                                                           k=0
                                                         =127 + (−1 + 40.735i)e  πi/2  + (−1 + 20.355i)e  πi
                                                           + (−1 + 13.557i)e 3πi/2  + (−1 + 10.153i)e 2πi
                                                           + (−1 + 8.1078i)e 5πi/2  + (−1 + 6.7415i)e 3πi
                                                           + (−1 + 5.7631i)e 7πi/2  + (−1 + 5.0273i)e 4πi
                                                           + (−1 + 4.4532i)e 9πi/2  + (−1 + 3.9922i)e 5πi

                                                           + (−1 − 3.9922i)e 118πi/2  + (−1 − 4.4532i)e 1119πi/2
                                                           + (−1 − 5.0273i)e 120πi/2  + (−1 − 5.7631i)e 121π/2
                                                           + (−1 − 6.7415i)e 122πi/2  + (−1 − 8.1078i)e 123πi/2
                                                           + (−1 − 10.153i)e 124πi/2  + (−1 − 13.557i)e 125πi/2
                                                           + (−1 − 20.355i)e 126πi/2  + (−1 − 30.735i)e 127πi/2
                                                         =61.04832.

                                        Then
                                                                       127
                                                                    1       πijk/64
                                                                         V k e   = 0.47694.
                                                                   128
                                                                       k=0
                                        This gives the 128 point DFT approximation of 0.47694 to the sampled partial sum S 10 (1/2),
                                        which was computed to be 0.45847. The difference is 0.0185.
                                           The actual sum of the Fourier series at t = 1/2is f (1/2) = 0.50000. In practice we would
                                        achieve greater accuracy by choosing N larger, allowing larger M.




                               SECTION 14.6        PROBLEMS


                            In each of Problems 1 through 6, a function is given  1. f (t) = 1 + t for 0 ≤ t < 2, p = 2, t 0 = 1/8
                                                                                   2
                            having period p. Compute the complex Fourier series  2. f (t) = t for 0 ≤ t < 1, p = 1, t 0 = 1/2
                            of the function and then the 10th partial sum of  3. f (t) = cos(t) for 0 ≤ t < 2, p = 2, t 0 = 1/8
                            this series at the indicated t 0 . Then, using N = 128,  4. f (t) = e −t  for 0 ≤ t < 4, p = 4, t 0 = 1/4
                                                                                   3
                            compute a DFT approximation to this partial sum  5. f (t) = t for 0 ≤ t < 1, p = 1, t 0 = 1/4
                            at t 0 .                                       6. f (t) = t sin(t) for 0 ≤ t < 4, p = 4, t 0 = 1/8.



                            14.7        DFT Approximation of the Fourier Transform

                                        Under certain conditions the DFT can be used to approximate the Fourier transform of a function.
                                                       ˆ
                                        To begin, suppose f (ω) can be approximated to within some acceptable tolerance by an integral
                                        over a finite interval
                                                                    ∞
                                                                                     2π L
                                                            ˆ
                                                            f (ω) =   f (ξ)e −iωξ  dξ ≈  f (ξ)e −iωξ  dξ.
                                                                   −∞              0

                      Copyright 2010 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
                      Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.

                                   October 14, 2010  16:43  THM/NEIL   Page-501        27410_14_ch14_p465-504
   516   517   518   519   520   521   522   523   524   525   526